Application of Near Infrared Spectra to Discriminating Honey Based on Partial Least Squares
The model of discrimination analysis was established by partial least squares (PLS)-Mahalanobois distance for differentiating raw honey and adulterated honey which was added with starch syrup. The effects of principal component number, pretreatment methods of spectroscopy and spectral region on the accuracy in discrimination of adulterated honey were studied. The results indicated that the cumulate reliabilities of the first seven principal components were 99.97%, and the highest recognition accuracy was offered by the model which was established with standard normalized variate near infrared spectra in 4000 cm-1-11493 cm-1 wavelength range. Twenty unknown samples were predicted by the model, and the discrimination rate is 100%.
Near infrared spectroscopy Honey Partial least squares adulterated honey
Xia Liya Li Honglian Zhao Zhilei
Supervision Institution of Quality&Technology, Hebei University, Baoding 071002
国际会议
The 6th International Conference on Partial Least Squares and Related Methods(第六届偏最小二乘及相关方法国际会议)
北京
英文
307-310
2009-09-04（万方平台首次上网日期，不代表论文的发表时间）